A further critique of the analytic strategy of adjusting for covariates to identify biologic mediation - PubMed (original) (raw)
A further critique of the analytic strategy of adjusting for covariates to identify biologic mediation
Jay S Kaufman et al. Epidemiol Perspect Innov. 2004.
Abstract
BACKGROUND: Epidemiologic research is often devoted to etiologic investigation, and so techniques that may facilitate mechanistic inferences are attractive. Some of these techniques rely on rigid and/or unrealistic assumptions, making the biologic inferences tenuous. The methodology investigated here is effect decomposition: the contrast between effect measures estimated with and without adjustment for one or more variables hypothesized to lie on the pathway through which the exposure exerts its effect. This contrast is typically used to distinguish the exposure's indirect effect, through the specified intermediate variables, from its direct effect, transmitted via pathways that do not involve the specified intermediates. METHODS: We apply a causal framework based on latent potential response types to describe the limitations inherent in effect decomposition analysis. For simplicity, we assume three measured binary variables with monotonic effects and randomized exposure, and use difference contrasts as measures of causal effect. Previous authors showed that confounding between intermediate and the outcome threatens the validity of the decomposition strategy, even if exposure is randomized. We define exchangeability conditions for absence of confounding of causal effects of exposure and intermediate, and generate two example populations in which the no-confounding conditions are satisfied. In one population we impose an additional prohibition against unit-level interaction (synergism). We evaluate the performance of the decomposition strategy against true values of the causal effects, as defined by the proportions of latent potential response types in the two populations. RESULTS: We demonstrate that even when there is no confounding, partition of the total effect into direct and indirect effects is not reliably valid. Decomposition is valid only with the additional restriction that the population contain no units in which exposure and intermediate interact to cause the outcome. This restriction implies homogeneity of causal effects across strata of the intermediate. CONCLUSIONS: Reliable effect decomposition requires not only absence of confounding, but also absence of unit-level interaction and use of linear contrasts as measures of causal effect. Epidemiologists should be wary of etiologic inference based on adjusting for intermediates, especially when using ratio effect measures or when absence of interacting potential response types cannot be confidently asserted.
Figures
Figure 1
Decomposition of Total Effect of X on Y into Direct and Indirect Effects. The total average causal effect (ACE) of X on Y is achieved through two pathways, one which is termed "indirect'' because it operates through measured intermediate variable Z, and another that is termed "direct'' because it operates through no measured intermediates.
Similar articles
- Estimation of direct causal effects.
Petersen ML, Sinisi SE, van der Laan MJ. Petersen ML, et al. Epidemiology. 2006 May;17(3):276-84. doi: 10.1097/01.ede.0000208475.99429.2d. Epidemiology. 2006. PMID: 16617276 - Improved estimation of controlled direct effects in the presence of unmeasured confounding of intermediate variables.
Kaufman S, Kaufman JS, MacLehose RF, Greenland S, Poole C. Kaufman S, et al. Stat Med. 2005 Jun 15;24(11):1683-702. doi: 10.1002/sim.2057. Stat Med. 2005. PMID: 15742358 - Toward a clearer understanding of causal concepts in epidemiology.
Maldonado G. Maldonado G. Ann Epidemiol. 2013 Dec;23(12):743-9. doi: 10.1016/j.annepidem.2013.09.001. Ann Epidemiol. 2013. PMID: 24404565 - Mediation analysis in epidemiology: methods, interpretation and bias.
Richiardi L, Bellocco R, Zugna D. Richiardi L, et al. Int J Epidemiol. 2013 Oct;42(5):1511-9. doi: 10.1093/ije/dyt127. Epub 2013 Sep 9. Int J Epidemiol. 2013. PMID: 24019424 Review. - Making valid causal inferences from observational data.
Martin W. Martin W. Prev Vet Med. 2014 Feb 15;113(3):281-97. doi: 10.1016/j.prevetmed.2013.09.006. Epub 2013 Sep 23. Prev Vet Med. 2014. PMID: 24113257 Review.
Cited by
- Methodologic Strategies for Quantifying Associations of Historical and Contemporary Mortgage Discrimination on Population Health Equity: A Systematic Review.
Edwards JR, Ong C, Barber S, Headen I, Tabb LP, De Roos AJ, Schinasi LH. Edwards JR, et al. J Racial Ethn Health Disparities. 2024 Sep 17. doi: 10.1007/s40615-024-02137-3. Online ahead of print. J Racial Ethn Health Disparities. 2024. PMID: 39289334 Review. - Women with hearing loss show increased dementia risk and brain atrophy.
Fitzhugh MC, Pa J. Fitzhugh MC, et al. Alzheimers Dement (Amst). 2023 Nov 27;15(4):e12499. doi: 10.1002/dad2.12499. eCollection 2023 Oct-Dec. Alzheimers Dement (Amst). 2023. PMID: 38026760 Free PMC article. - Reevaluating the protective effect of smoking on preeclampsia risk through the lens of bias.
Rodriguez-Lopez M, Escobar MF, Merlo J, Kaufman JS. Rodriguez-Lopez M, et al. J Hum Hypertens. 2023 May;37(5):338-344. doi: 10.1038/s41371-023-00827-9. Epub 2023 Apr 11. J Hum Hypertens. 2023. PMID: 37041252 Free PMC article. Review. - Insulin resistance mediates obesity-related risk of cardiovascular disease: a prospective cohort study.
Tian X, Chen S, Wang P, Xu Q, Zhang Y, Luo Y, Wu S, Wang A. Tian X, et al. Cardiovasc Diabetol. 2022 Dec 23;21(1):289. doi: 10.1186/s12933-022-01729-9. Cardiovasc Diabetol. 2022. PMID: 36564775 Free PMC article. - Trends for the Association between Body Mass Index and Risk of Cardiovascular Disease among the Japanese Population: The Circulatory Risk in Communities Study (CIRCS).
Matsumura T, Sankai T, Yamagishi K, Tanaka M, Kubota Y, Hayama-Terada M, Shimizu Y, Muraki I, Umesawa M, Cui R, Imano H, Ohira T, Kitamura A, Okada T, Kiyama M, Iso H. Matsumura T, et al. J Atheroscler Thromb. 2023 Apr 1;30(4):335-347. doi: 10.5551/jat.63415. Epub 2022 Jul 28. J Atheroscler Thromb. 2023. PMID: 35896353 Free PMC article.
References
- Koopman JS, Weed DL. Epigenesis theory: a mathematical model relating causal concepts of pathogenesis in individuals to disease patterns in populations. Am J Epidemiol. 1990;132:366–390. - PubMed
- Greenland S, Poole C. Invariants and noninvariants in the concept of interdependent effects. Scand J Work Environ Health. 1988;14:125–129. - PubMed
- Greenland S, Rothman KJ. Concepts of Interaction. In: Rothman KJ, Greenland S, editor. Modern Epidemiology. 2. Philadelphia, Pa.: Lippincott-Raven; 1998. pp. 329–342.
- MacLehose RF, Kaufman S, Kaufman JS, Poole C. Bounding causal effects under uncontrolled confounding using counterfactuals. Epidemiology. - PubMed
LinkOut - more resources
Full Text Sources